Summary: Group B Streptococci (GBS) are the leading cause of neonatal sepsis and meningitis. Maternal IgG antibodies to the GBS polysaccharide (PS) protect the neonate from invasive GBS disease. Experimental conjugate vaccines have therefore been produced using reductive amination. We investigated effects of different conjugation methods on the epitopes and immunological reactivity of the GBS type III PS by ELISA and inhibition ELISA. We compared the reductive amination and cyanylation conjugation methods for their effects on immunological reactivity of the GBS type III PS. Monoclonal antibodies (Mab) to GBS type III and to the closely related Pn 14 PS were used to investigate possible alterations in epitope availability or exposure during PS activation and conjugation. Binding of a human post-PS immunization serum, S19, and several different Mabs, all at 50 ng/ml of IgG anti-PS antibody was examined. The S19 antibody bound to native GBS III PS epitopes not altered by reductive amination, and was used to set the antigen coating conditions such that S19 binding was similar for each of the GBS III antigens. GBS III Mabs, 53-1 and 54-1, bound to native and modified GBS III antigens, but neither bound to the Pn 14 PS. The Pn 14 Mab, 42-1, showed negligible reactivity with the native GBS III PS and two GBS conjugate antigens prepared by cyanylation. In contrast, this Mab reacted strongly with two GBS III conjugates prepared by reductive amination. Additional studies showed that periodate oxidation, the first step in reductive amination, of the native GBS type III PS was sufficient to expose an epitope recognized by some Pn 14 Mabs. These Mabs were opsonic for Pn 14 but not for GBS type III. Importantly, as predicted by the Mab data, we found that the altered GBS III epitope induces Pn 14 reactive antibodies in humans that were only opsonic for Pn 14. The Pn 14 Mab data show that the native type 14 PS has at least two distinct epitopes, 14a and 14b. The 14a epitope is the serotype 14 specific antigen, while the 14b epitope is the cross-reactive epitope found on both the type 14 PS and chemically modified GBS III PS. Since neither Mab was opsonic for GBS type III, the 14b epitope is not accessible on GBS III cells. These Mab data suggest two potentially important issues regarding clinical immunogenicity. First, if the 14b epitope is present in the GBS III-tetanus toxoid conjugate and is immunogenic, then there is a probability that some GBS conjugate vaccinees will respond predominately to the 14b epitope. In fact, 20% of the GBS III-tetanus toxoid vaccinees that we evaluated had a 3-fold or greater antibody rise post immunization to a GBS III-HSA conjugate as well as to the Pn 14 PS, with little or no increase to the native GBS III PS. Furthermore, these sera were not opsonic for GBS III, yet were opsonic for Pn 14; all consistent with our Mab data.